bulkWrite() method

db.collection.bulkWrite() method can perform these operations:

• insertOne
• updateOne
• updateMany
• replaceOne
• deleteOne
• deleteMany

It has the following syntax:

db.collection(
    [
        {
            <operation1>: {
                "<option1>": <document1>,
                "<option2>": <document2>
            }
        },
        {
            <operation2>: {
                "<option1>": <document1>,
                "<option2>": <document2>
            }
        },
        ...
    ],
    {
        writeConcern: <document>,
        ordered: <boolean>
    }
)

bulkWrite() method can be either ordered or unordered, meaning the array of operations can be executed in series or in parallel. By default, MongoDB performs bulkWrite() in ordered mode.

For example:

db.customers.bulkWrite(
    [
        {
            insertOne: {
                "document": {
                    email: "[email protected]",
                    age: 210,
                    gender: "M"
                }
            }
        },
        {
            insertOne: {
                "document": {
                    email: "[email protected]",
                    age: 220,
                    gender: "F"
                }
            }
        },
        {
            updateMany: {
                "filter": {
                    age: {
                        $gt: 100
                    }
                },
                "update": {
                    $set: {
                        age: 10
                    }
                }
            }
        },
        {
            replaceOne: {
                "filter": {
                    email: "[email protected]"
                },
                "replacement": {
                    email: "[email protected]",
                    age: 10,
                    gender: "F"
                }
            }
        },
        {
            deleteMany: {
                "filter": {
                    age: {
                        $eq: 10
                    }
                }
            }
        }
    ]
)

In this example:

• Add two documents
• Update all documents with age >= 200, and set their age to 10
• Replace a document matching email [email protected] with a new document
• Remove all documents that has age equal to 10

Output:

{
  acknowledged: true,
  insertedCount: 2,
  insertedIds: {
    '0': ObjectId("60bcbeedd483412bb240775d"),
    '1': ObjectId("60bcbeedd483412bb240775e")
  },
  matchedCount: 3,
  modifiedCount: 3,
  deletedCount: 2,
  upsertedCount: 0,
  upsertedIds: {}
}

We can also use aggregation pipelines inside the updateOne or updateMany operations:

db.customers.bulkWrite(
    [
        {
            insertOne: {
                "document": {
                    email: "[email protected]",
                    age: 210,
                    gender: "M"
                }
            }
        },
        {
            updateMany: {
                "filter": {
                    age: {
                        $gt: 200
                    }
                },
                "update": [
                    {
                        $set: {
                            age: 10
                        }
                    },
                    {
                        $project: {
                            email: 1,
                            age: 1
                        }
                    }
                ]
            }
        },
        {
            deleteMany: {
                "filter": {
                    age: {
                        $eq: 10
                    }
                }
            }
        }
    ]
)

Output:

{
  acknowledged: true,
  insertedCount: 1,
  insertedIds: { '0': ObjectId("60bcc32ed483412bb2407767") },
  matchedCount: 1,
  modifiedCount: 1,
  deletedCount: 1,
  upsertedCount: 0,
  upsertedIds: {}
}